Skip to content

AzusaOS/cudapreload

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

cudapreload

This library will allow most tools (torch, etc) needing access to nvidia drivers to work fine. This will analyze the currently loaded nvidia kernel driver and direct dlopen() calls to the right libraries, meaning that even if you upgraded recently but still have an old kernel driver loaded it will still work fine.

Usage

For example with an empty venv

$ /pkg/main/dev-lang.python.core/bin/python -m venv test
$ cd test
$ . bin/activate
$ pip install torch numpy
[...]
$ python -c 'import torch; print(torch.cuda.is_available())'
False
$ LD_PRELOAD=/pkg/main/sys-libs.cudapreload.dev/lib64/cudapreload.so python -c 'import torch; print(torch.cuda.is_available())'
True

About

Preload library that can be used to load cuda from its path outside of Azusa

Resources

License

Stars

Watchers

Forks

Packages

No packages published